Associations of built environmental characteristics and physical activity in neighbourhoods of Accra, Ghana.

Residents of high residential density and high socioeconomic status neighborhoods had the highest levels of physical activity, mainly due to work-related physical activity.

• A cross-sectional study design was used in Accra, Ghana.
• 12 neighborhoods were randomly selected based on high and low values of residential density and socioeconomic status.
• Physical activity was assessed with the modified WHO Global Physical Activity Questionnaire.
• Work-related physical activity was identified as the primary driver of higher activity levels in high-density, high-SES neighborhoods.

Closer proximity of destinations and recreational facilities was associated with increased physical activity levels among residents.

• Built environment characteristics were assessed using the Neighborhood Environment Walkability Scale questionnaire adapted for Africa (NEWS-A).
• Generalized linear models were performed to determine associations between neighborhood characteristics and physical activity.
• Proximity of both destinations and recreational facilities were independently identified as positive correlates of physical activity.

Greater street connectivity, better aesthetics, greater safety when walking and cycling, and safety from crime were associated with increased physical activity levels.

• These built environment characteristics were measured via the NEWS-A questionnaire.
• Multiple built environment attributes were significantly associated with physical activity in the generalized linear model analyses.
• Both traffic safety (walking and cycling) and crime safety were identified as distinct positive correlates of physical activity.

Social cohesion demonstrated no significant association with physical activity.

• Social cohesion was measured using the Perceived Neighborhood Social Cohesion Questionnaire.
• Unlike built environment characteristics, social cohesion did not reach statistical significance as a correlate of physical activity.
• This finding contrasts with some Western-country literature that has linked social cohesion to physical activity.

There were significant differences in built environment characteristics across different neighborhood types in Accra.

• Neighborhoods were categorized by combinations of high/low residential density and high/low socioeconomic status.
• 12 neighborhoods were randomly selected to represent variation in these dimensions.
• Sociodemographic, household, and neighborhood information was collected alongside built environment data.
• The study contributes data from a low and middle-income country context where such evidence has been limited.